New approaches for intervention in virally-induced disease are always sought after. Interest has been shown in adapting gene therapy for use in delivery of DNA immunogens to target cells and tissues. Success in this endeavour has been limited, however, partly because selective delivery of genes to target tissues has proven to be extremely difficult, and partly because success is dependent upon being able to predictably elicit either a specific antibody response or a cell-mediated response or both. Which response is desired, depends on the nature of the presenting problem.
While use of tissue-specific promoters to target gene therapy has been effective in some animal models, it has proven less so in man and selective tissue-specific promoters are not available for a wide range of tissues. Gene delivery systems such as vectors are not themselves tissue selective. Viral vectors, such as retroviruses and adenovirus, may be used and are to some extent selective. However, many vector systems are by their nature unable to produce stable integrants. Moreover, some also invoke immune responses themselves, thereby precluding predictable and effective immunological intervention.
A further reason why gene therapy has proven to be difficult is the unpredictability of the kind of immune response which can be generated as a result of delivery of packaged DNA, by whatever means, to a cell or tissue. For example, it is known that DNA vaccines delivered to the skin tend to cause the induction of antibodies, whereas the same vaccine delivered intramuscularly may favour a cellular response. In instances where virus infection is associated with long latency periods, such as is the case, for example, with infection by human immunodeficiency virus and herpes simplex virus, it may be desirable to be able to elicit responses from both arms of the immune system to allow simultaneous protection against future infection and treatment of existing infection.
One example where the unpredictability of the response is of concern is in attempts to overcome the effects of infection by the human papilloma virus (HPV). Many genotypes of HPV are known to be antecedent to ano-genital malignancy1; tumours can develop after a latency period of from 10 to 30 years. There is a need, therefore, to find means to induce both neutralising antibody to prevent HPV infection, and cell-mediated immunity to treat existing infection. Induction of a cell-mediated immune response to viral genes is desirable both to control existing infection and, particularly, to eliminate the premalignant consequences of HPV infection. A vaccine, which was capable of effecting both humoral and cellular immunity, would constitute a major advance in medicine's capability to intervene effectively in such disease conditions.
Recent attempts to increase the level of cellular response following DNA immunisation have had some success through conjugation of the sequences encoding the cellular protein ubiquitin. DNA immunisation as a form of gene therapy is a relatively new approach to vaccination. In theory, and sometimes in practice, the inoculated plasmid DNA enters the cell and the encoded proteins are expressed therein. In this way, access of the antigen to the major histocompatibility complex (C) class I antigen presentation pathway is more likely to occur. In some cases, enhanced Cytotoxic T Lymphocyte (CTL) responses have been shown following ubiquitination31,32. However, ubiquitination is known not to enhance the immunogenicity of all polynucleotide vaccines35. Furthermore, the increased CTL response is sometimes at the expense of immunogenicity and production of neutralising antibody33.
Furthermore, the present inventors have previously shown that the L1 major capsid protein of HPV can self-assemble into virus-like particles (VLPs)3. Used as a vaccine, VLPs elicit conformation-dependent host-protective neutralising antibody4. However, papilloma virus (PV) infection is often persistent, and 20-50% of HPV-infected individuals do not develop PV-specific immunity5-7, suggesting that PV infection is seen poorly by the immune system8. The capsid genes of HPV have been shown to have sub-optimal codon usage for expression in mammalian cells12.